Selection of optimal exciting frequency and Lamb wave mode for detecting wall-thinning in plates using scanning laser Doppler vibrometer. (15th August 2022)

• Record Type: Journal Article
• Authors: Kang, To; Yoon, Minho; Han, Soonwoo; Kim, Kyung-Mo; Koo, Bonyong
• Is Part Of: Measurement. Volume 200(2022)
• Publication Date: 2022-08-15
• Subjects: Optimization -- IDT-based SLDV -- Lamb wave -- Wavenumber sensitivity -- Minimum distance from reference mode (MDRM); NDE non-destructive evaluation -- SLDV scanning laser Doppler vibrometer -- IDT interdigital transducer -- MDRM minimum distance from reference mode -- UPI ultrasonic propagation imager -- AWS acoustic wave spectroscopy -- 2D-FFT two-dimensional fast Fourier transform -- DAQ data acquisition; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2022.111676
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: The optimal Lamb wave mode and frequency were selected through an optimization process to detect wall thinning in the plate efficiently. The optimization problem is formulated to minimize the number of inspection modes with constraints in wavenumber sensitivity and minimum distance from the reference mode. The selected optimal frequencies and modes were experimentally validated using an IDT-based SLDV. Abstract: The continuous excitation-based scanning laser Doppler vibrometer (SLDV) has been gaining prominence as a defect-detecting tool because of its superior performance in discovering faults, such as thinning of structures, compared to conventional non-destructive evaluation methods. To date, the optimal test mode and frequency for interdigital transducers have been heuristically selected. Although this approach is effective for problems involving a small number of modes, heuristic algorithms have limitations with respect to problems involving multiple complex modes. Therefore, in this study, the determination of the inspection mode and frequency is formulated as an optimization problem, and the effective wavenumber and bandwidth of the high-pass and bandpass filters in signal processing are also considered. Numerical optimization examples show that the proposed method is effective and robust. An experiment with a thick plate with shallow defects was performed, and the quantitative images obtained showed the correct location of the defects with the proposed numerical optimal solution.